About Me

• I’m an Assistant Professor in the College of Education
  • I enjoy model building, particularly longitudinal models, and statistical programming.
• I’ve used R for over 10 years.
  • I have 4 R packages, 3 on CRAN, 1 on GitHub
    • simglm
    • pdfsearch
    • highlightHTML
    • SPSStoR
• GitHub Repository for this workshop: https://github.com/lebebr01/iowa_ds_graphics

Why teach the tidyverse

• The tidyverse is a series of packages developed by Hadley Wickham and his team at RStudio. https://www.tidyverse.org/
• I teach/use the tidyverse for 3 major reasons:
  • Simple functions that do one thing well
  • Consistent implementations across functions within tidyverse (i.e. common APIs)
  • Provides a framework for data manipulation

Course Setup

install.packages("tidyverse")

library(tidyverse)

## Registered S3 methods overwritten by 'ggplot2':
##   method         from 
##   [.quosures     rlang
##   c.quosures     rlang
##   print.quosures rlang

## Registered S3 method overwritten by 'rvest':
##   method            from
##   read_xml.response xml2

## ── Attaching packages ──────────────────────────────────────────────────────────────────────── tidyverse 1.2.1 ──

## ✔ ggplot2 3.1.1       ✔ purrr   0.3.2  
## ✔ tibble  2.1.1       ✔ dplyr   0.8.0.1
## ✔ tidyr   0.8.3       ✔ stringr 1.4.0  
## ✔ readr   1.3.1       ✔ forcats 0.4.0

## ── Conflicts ─────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()

Explore Data

PID county state area poptotal popdensity popwhite popblack 1 561 ADAMS IL 0.052 66090 1270.9615 63917 1702 2 562 ALEXANDER IL 0.014 10626 759.0000 7054 3496 3 563 BOND IL 0.022 14991 681.4091 14477 429 4 564 BOONE IL 0.017 30806 1812.1176 29344 127 5 565 BROWN IL 0.018 5836 324.2222 5264 547 6 566 BUREAU IL 0.050 35688 713.7600 35157 50 popamerindian popasian popother percwhite percblack percamerindan 1 98 249 124 96.71206 2.5752761 0.1482826 2 19 48 9 66.38434 32.9004329 0.1788067 3 35 16 34 96.57128 2.8617170 0.2334734 4 46 150 1139 95.25417 0.4122574 0.1493216 5 14 5 6 90.19877 9.3728581 0.2398903 6 65 195 221 98.51210 0.1401031 0.1821340 percasian percother popadults perchsd percollege percprof 1 0.37675897 0.18762294 43298 75.10740 19.63139 4.355859 2 0.45172219 0.08469791 6724 59.72635 11.24331 2.870315 3 0.10673071 0.22680275 9669 69.33499 17.03382 4.488572 4 0.48691813 3.69733169 19272 75.47219 17.27895 4.197800 5 0.08567512 0.10281014 3979 68.86152 14.47600 3.367680 6 0.54640215 0.61925577 23444 76.62941 18.90462 3.275891 poppovertyknown percpovertyknown percbelowpoverty percchildbelowpovert 1 63628 96.27478 13.151443 18.01172 2 10529 99.08714 32.244278 45.82651 3 14235 94.95697 12.068844 14.03606 4 30337 98.47757 7.209019 11.17954 5 4815 82.50514 13.520249 13.02289 6 35107 98.37200 10.399635 14.15882 percadultpoverty percelderlypoverty inmetro category 1 11.009776 12.443812 0 AAR 2 27.385647 25.228976 0 LHR 3 10.852090 12.697410 0 AAR 4 5.536013 6.217047 1 ALU 5 11.143211 19.200000 0 AAR 6 8.179287 11.008586 0 AAR

First ggplot

ggplot(data = midwest) +
  geom_point(mapping = aes(x = popdensity, y = percollege))

Equivalent Code

ggplot(midwest) +
  geom_point(aes(x = popdensity, y = percollege))

Your Turn

1. Try plotting popdensity by state.
2. Try plotting county by state.
   • Does this plot work?
3. Bonus: Try just using the ggplot(data = midwest) from above.
   • What do you get?
   • Does this make sense?

Add Aesthetics

ggplot(midwest) +
  geom_point(aes(x = popdensity, y = percollege, color = state))

Global Aesthetics

ggplot(midwest) +
  geom_point(aes(x = popdensity, y = percollege), color = 'pink')

Your Turn

1. Instead of using colors, make the shape of the points different for each state.
2. Instead of color, use alpha instead.
   • What does this do to the plot?
3. Try the following command: colors().
   • Try a few colors to find your favorite.
4. What happens if you use the following code:

ggplot(midwest) + 
  geom_point(aes(x = popdensity, y = percollege, color = 'green'))

Additional Geoms

ggplot(midwest) +
  geom_smooth(aes(x = popdensity, y = percollege))

Add more Aesthetics

ggplot(midwest) +
  geom_smooth(aes(x = popdensity, y = percollege, linetype = state), 
              se = FALSE)

Your Turn

1. It is possible to combine geoms, which we will do next, but try it first. Try to recreate this plot.

Layered ggplot

ggplot(midwest) +
  geom_point(aes(x = popdensity, y = percollege, color = state)) +
  geom_smooth(aes(x = popdensity, y = percollege, color = state), 
              se = FALSE)

Remove duplicate aesthetics

ggplot(midwest, 
       aes(x = popdensity, y = percollege, color = state)) +
  geom_point() +
  geom_smooth(se = FALSE)

Your Turn

1. Can you recreate the following figure?

Brief plot customization

ggplot(midwest, 
       aes(x = popdensity, y = percollege, color = state)) +
  geom_point() + 
  scale_x_continuous("Population Density", 
                     breaks = seq(0, 80000, 20000)) + 
  scale_y_continuous("Percent College Graduates") + 
  scale_color_discrete("State")

Brief plot customization Output

Change plot theme

ggplot(midwest, 
       aes(x = popdensity, y = percollege, color = state)) +
  geom_point() +
  geom_smooth(se = FALSE) + 
  theme_bw()

More themes

• Themes in ggplot2: http://ggplot2.tidyverse.org/reference/ggtheme.html
• Themes from ggthemes package: https://cran.r-project.org/web/packages/ggthemes/vignettes/ggthemes.html

Base plot for reference

p1 <- ggplot(midwest, 
       aes(x = popdensity, y = percollege, color = state)) +
  geom_point() + 
  scale_x_continuous("Population Density", 
                     breaks = seq(0, 80000, 20000)) + 
  scale_y_continuous("Percent College Graduates") + 
  theme_bw()

Add plot title or subtitle

p1 + 
  labs(title = "Percent College Educated by Population Density",
       subtitle = "County level data for five midwest states")

Color Options

p1 + scale_color_grey("State")

Using colorbrewer2.org

• http://colorbrewer2.org

p1 + scale_color_brewer("State", palette = 'Dark2')

Two additional color options

• viridis: https://github.com/sjmgarnier/viridis
• scico: https://github.com/thomasp85/scico

viridis colors

library(viridis)
p1 + scale_color_viridis(discrete = TRUE)

viridis colors

p1 + scale_color_viridis(option = 'cividis', discrete = TRUE)

Zoom in on a plot

ggplot(data = midwest, 
       aes(x = popdensity, y = percollege, color = state)) +
  geom_point() + 
  scale_x_continuous("Population Density") + 
  scale_y_continuous("Percent College Graduates") + 
  scale_color_discrete("State") + 
  coord_cartesian(xlim = c(0, 15000))

Zoom in on a plot output

Zoom using scale_x_continuous - Bad Practice

ggplot(data = midwest, 
       aes(x = popdensity, y = percollege, color = state)) +
  geom_point() + 
  geom_smooth(se = FALSE) +
  scale_x_continuous("Population Density", limits = c(0, 15000)) + 
  scale_y_continuous("Percent College Graduates") + 
  scale_color_discrete("State")

Comparing output

## Warning: Removed 16 rows containing non-finite values (stat_smooth).

## Warning: Removed 16 rows containing missing values (geom_point).

Lord of the Rings Data

• Data from Jenny Bryan: https://github.com/jennybc/lotr

lotr <- read_tsv('https://raw.githubusercontent.com/jennybc/lotr/master/lotr_clean.tsv')

## Parsed with column specification:
## cols(
##   Film = col_character(),
##   Chapter = col_character(),
##   Character = col_character(),
##   Race = col_character(),
##   Words = col_double()
## )

View LOTR

head(lotr)

## # A tibble: 6 x 5
##   Film                       Chapter                Character Race   Words
##   <chr>                      <chr>                  <chr>     <chr>  <dbl>
## 1 The Fellowship Of The Ring 01: Prologue           Bilbo     Hobbit     4
## 2 The Fellowship Of The Ring 01: Prologue           Elrond    Elf        5
## 3 The Fellowship Of The Ring 01: Prologue           Galadriel Elf      460
## 4 The Fellowship Of The Ring 02: Concerning Hobbits Bilbo     Hobbit   214
## 5 The Fellowship Of The Ring 03: The Shire          Bilbo     Hobbit    70
## 6 The Fellowship Of The Ring 03: The Shire          Frodo     Hobbit   128

Geoms for single variables

ggplot(lotr, aes(x = Words)) + 
  geom_histogram() + 
  theme_bw()

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

Customize histogram

ggplot(lotr, aes(x = Words)) + 
  geom_histogram(bins = 20) + 
  theme_bw()

Customize histogram 2

ggplot(lotr, aes(x = Words)) + 
  geom_histogram(binwidth = 25) + 
  theme_bw()

Histograms by other variables - likely not useful

ggplot(lotr, aes(x = Words, color = Film)) + 
  geom_histogram(binwidth = 25) + 
  theme_bw()

Histograms by other variables - one alternative

ggplot(lotr, aes(x = Words)) + 
  geom_histogram(binwidth = 25) + 
  theme_bw() + 
  facet_wrap(~ Film)

Your Turn

1. With more than two groups, histograms are difficult to interpret due to overlap. Instead, use the geom_density to create a density plot for Words for each film.
2. Using geom_boxplot, create boxplots with Words as the y variable and Film as the x variable. Bonus: facet this plot by the variable Race. Bonus2: Zoom in on the bulk of the data.

Rotation of axis labels

ggplot(lotr, aes(x = Film, y = Words)) + 
  geom_boxplot() + 
  facet_wrap(~ Race) + 
  theme_bw() + 
  theme(axis.text.x = element_text(angle = 90))

Many times coord_flip is better

ggplot(lotr, aes(x = Film, y = Words)) + 
  geom_boxplot() + 
  facet_wrap(~ Race) + 
  theme_bw() + 
  coord_flip()

Bar graphs

ggplot(lotr, aes(x = Race)) + 
  geom_bar() + 
  theme_bw()

Add aesthetic

ggplot(lotr, aes(x = Race)) +
  geom_bar(aes(fill = Film)) + 
  theme_bw()

Stacked Bars Relative

ggplot(lotr, aes(x = Race)) + 
  geom_bar(aes(fill = Film), position = 'fill') + 
  theme_bw() + 
  ylab("Proportion")

Dodged Bars

ggplot(lotr, aes(x = Race)) + 
  geom_bar(aes(fill = Film), position = 'dodge') + 
  theme_bw()

Change Bar Col bar_coloror

ggplot(lotr, aes(x = Race)) + 
  geom_bar(aes(fill = Film), position = 'fill') + 
  theme_bw() + 
  ylab("Proportion") + 
  scale_fill_viridis(option = 'cividis', discrete = TRUE)

Your Turn

1. Using the gss_cat data, create a bar chart of the variable partyid.
2. Add the variable marital to the bar chart created in step 1. Do you prefer a stacked or dodged version?
3. Take steps to make one of the plots above close to publication quality.

Additional ggplot2 resources

• ggplot2 website: http://docs.ggplot2.org/current/index.html
• ggplot2 book: http://www.springer.com/us/book/9780387981413
• R graphics cookbook: http://www.cookbook-r.com/Graphs/

Additional R Resources

• R for Data Science: http://r4ds.had.co.nz/

Why Interactive Graphics

• Why interactive graphics?
  • Created specifically for the web.
  • Can focus, explore, zoom, or remove data at will.
  • Allows users to customize their experience.
  • It is fun!

Interactive graphics with plotly

install.packages("plotly")

First Interactive Plot

library(plotly)
p <- ggplot(data = midwest) +
  geom_point(mapping = aes(x = popdensity, y = percollege))
ggplotly(p, width = 900, height = 500)

Customized Plot

p <- ggplot(midwest, 
       aes(x = popdensity, y = percollege, color = state)) +
  geom_point() + 
  scale_x_continuous("Population Density", 
                     breaks = seq(0, 80000, 20000)) + 
  scale_y_continuous("Percent College Graduates") + 
  scale_color_discrete("State") + 
  theme_bw()

Interactive Output

ggplotly(p, width = 900)

Your Turn

1. Using the starwars data, create a static ggplot and use the ggplotly function to turn it interactive.

Create plotly by hand

plot_ly(lotr, x = ~Words, width = 900) %>% add_histogram()

Subplots Code

one_plot <- function(d) {
  plot_ly(d, x = ~Words, width = 900) %>%
    add_histogram() %>%
    add_annotations(
      ~unique(Film), x = 0.5, y = 1, 
      xref = "paper", yref = "paper", showarrow = FALSE
    )
}

plotly_out <- lotr %>%
  split(.$Film) %>%
  lapply(one_plot) %>% 
  subplot(nrows = 1, shareX = TRUE, titleX = FALSE) %>%
  hide_legend()

Subplots Output

plotly_out

Grouped bar plot

plot_ly(lotr, x = ~Race, color = ~Film, width = 900) %>% add_histogram()

Plot of proportions code

## number of diamonds by cut and clarity (n)
lotr_count <- count(lotr, Race, Film)
## number of diamonds by cut (nn)
lotr_prop <- left_join(lotr_count, count(lotr_count, Race, wt = n), 
                       by = 'Race')

plotly_out <-lotr_prop %>%
  mutate(prop = n.x / n.y) %>%
  plot_ly(x = ~Race, y = ~prop, color = ~Film, width = 900) %>%
  add_bars() %>%
  layout(barmode = "stack")

Plot of proportions output

plotly_out

Your Turn

1. Using the gss_cat data, create a histrogram for the tvhours variable.
2. Using the gss_cat data, create a bar chart showing the partyid variable by the marital status.

Scatterplots by Hand

plot_ly(midwest, x = ~popdensity, y = ~percollege, width = 900) %>%
  add_markers()

Change symbol

plot_ly(midwest, x = ~popdensity, y = ~percollege, width = 900) %>%
  add_markers(symbol = ~state)

Change color

plot_ly(midwest, x = ~popdensity, y = ~percollege, width = 900) %>%
  add_markers(color = ~state, colors = viridis::viridis(5))

Line Graph

storms_yearly <- storms %>%
  group_by(year) %>%
  summarise(num = length(unique(name)))

plot_ly(storms_yearly, x = ~year, y = ~num, width = 900, height = 500) %>%
  add_lines()

Your Turn

1. Using the gss_cat data, create a scatterplot showing the age and tvhours variables.
2. Compute the average time spent watching tv by year and marital status. Then, plot the average time spent watching tv by year and marital status.

Highcharter; Highcharts for R

devtools::install_github("jbkunst/highcharter")

Load highcharter

library(highcharter)

## Registered S3 method overwritten by 'xts':
##   method     from
##   as.zoo.xts zoo

## Registered S3 method overwritten by 'quantmod':
##   method            from
##   as.zoo.data.frame zoo

## Highcharts (www.highcharts.com) is a Highsoft software product which is

## not free for commercial and Governmental use

lotr_count <- lotr %>%
  count(Film, Race)
head(lotr_count)

## # A tibble: 6 x 3
##   Film                       Race       n
##   <chr>                      <chr>  <int>
## 1 The Fellowship Of The Ring Dwarf     11
## 2 The Fellowship Of The Ring Elf       31
## 3 The Fellowship Of The Ring Hobbit   103
## 4 The Fellowship Of The Ring Man       40
## 5 The Fellowship Of The Ring Orc        3
## 6 The Fellowship Of The Ring Wizard    29

hchart function

hchart(lotr_count, "column", hcaes(x = Race, y = n, group = Film))

A second hchart

hchart(midwest, "scatter", hcaes(x = popdensity, y = percollege, group = state))

Histogram

hchart(lotr$Words)

Your Turn

1. Using the hchart function, create a bar chart or histogram with the gss_cat data.
2. Using the hchart function, create a scatterplot with the gss_cat data.

Build Highcharts from scratch

hc <- highchart() %>%
  hc_xAxis(categories = lotr_count$Race) %>%
  hc_add_series(name = 'The Fellowship Of The Ring', 
                data = filter(lotr_count, Film == 'The Fellowship Of The Ring')$n) %>% 
  hc_add_series(name = 'The Two Towers', 
                data = filter(lotr_count, Film == 'The Two Towers')$n) %>%
  hc_add_series(name = 'The Return Of The King', 
                data = filter(lotr_count, Film == 'The Return Of The King')$n)

Build highcharts from scratch output

hc

Change Chart type

hc <- hc %>%
  hc_chart(type = 'column')
hc

Change Colors

hc <- hc %>%
  hc_colors(substr(viridis(3), 0, 7))
hc

Modify Axes

hc <- hc %>%
  hc_xAxis(title = list(text = "Race")) %>%
  hc_yAxis(title = list(text = "Number of Words Spoken"),
           showLastLabel = FALSE)
hc

Add title, subtitle, move legend

hc <- hc %>%
  hc_title(text = 'Number of Words Spoken in Lord of the Rings Films',
           align = 'left') %>%
  hc_subtitle(text = 'Broken down by <i>Film</i> and <b>Race</b>', 
              align = 'left') %>%
  hc_legend(align = 'right', verticalAlign = 'top', layout = 'vertical',
            x = 0, y = 80) %>%
  hc_exporting(enabled = TRUE)

Add title, subtitle, move legend output

hc

Your Turn

1. Build up a plot from scratch, getting the figure close to publication quality using the gss_cat data.

Correlation Matrices

select(storms, wind, pressure, ts_diameter, hu_diameter) %>%
  cor(use = "pairwise.complete.obs") %>%
  hchart()

Leaflet Example

library(leaflet)

storms %>%
  filter(name %in% c('Ike', 'Katrina'), year > 2000) %>%
  leaflet() %>%
  addTiles() %>%
  addCircles(lng = ~long, lat = ~lat, popup = ~name, weight = 1,
             radius = ~wind*1000)

gganimate

install.packages("gganimate")

gganimate example

library(gganimate)

ggplot(storms, aes(x = pressure, y = wind, color = status)) + 
  geom_point(show.legend = FALSE) + 
  xlab("Pressure") + 
  ylab("Wind Speed (MPH)") +
  facet_wrap(~status) +
  theme_bw(base_size = 14) +
  labs(title = 'Year: {frame_time}') +
  transition_time(as.integer(year)) + 
  ease_aes('linear')

gganimate output

Additional Resources

• plotly for R book: https://plotly-book.cpsievert.me/
• plotly: https://plot.ly/
• highcharter: http://jkunst.com/highcharter/index.html
• highcharts: https://www.highcharts.com/
• htmlwidgets: https://www.htmlwidgets.org/
• gganimate: https://gganimate.com/